The long term goal of this project is to minimize organ rejection and immunosuppressant toxicity, in each child with liver transplantation (LTx). Pre-LTx lymphocyte depletion permits steroid avoidance and lowers need for Tacrolimus immunosuppression. If underlying mechanisms were better understood, they could be used to reduce further, primary rejection (50%) and recurrent rejection during drug minimization (30%) on this protocol. Preliminary work leads us to hypothesize that donor-specific hyporeactivity and regulatory- suppressive effect are achieved at highly variable intervals among pediatric LTx with early rejection after steroid-free lymphocyte-depleting immunosuppression. We further hypothesize that this variability in immune-modulation is associated with patterns of single nucleotide polymorphisms (SNP) in extended MHC- region genes subserving proliferation, apoptosis, memory and B-cell-dependent functions. A clinical trial at our center will administer steroid-free Tacrolimus after depletion with 5 mg/kg rabbit anti-human-thymocyte globulin (rATG) to 80 children with LTx. The proposed mechanistic addendum study will entail serial peripheral blood samples from all children before and at 1, 3, and 12 months post-LTx. Specific aims are 1. Longitudinal characterization of donor-specific alloreactivity, T-reg/suppressor cells (CD4+CD25+, CD8+28-), and anti-HLA alto-antibodies in each child, 2. Pre-LTx characterization of 29 SNPs distributed among 14 MHC genes, whose preliminary distribution patterns differ significantly between rejectors (biopsy-proven rejection at 60 days post-LTx), and non-rejectors. This will be done in 80 children and their biologic parents. SNPs showing >or <50% expected transmission from parents to rejectors and whose transmission differs significantly between rejectors and non-rejectors, will be used to identify candidate loci with the transmission disequilibrium test, and 3. Validate potential candidate genes/loci within outcome groups, by a) measuring donor-specific proliferative/apoptotic/memory responses in T- and B-cell subsets by CFSE-MLR, b) whole genome mRNA expression to complement SNP associations, and minimize false-positives, and c) locus- specific gene expression (mRNA) for candidate loci c). If successful, future application of study results could improve pre-LTx drug selection, e.g.,allocating steroids if SNP patterns predict rejection. Also, post-LTx drug minimization could be made safer, e.g., when T-reg/T-sup appear.